Search

Hierarchical task decomposition is a method used in many agent systems to organize agent knowledge. This work shows how the combination of a hierarchy and persistent assertions of knowledge can lead to difficulty in maintaining logical consistency in asserted knowledge. We explore the problematic consequences of persistent assumptions in the reasoning process and introduce novel potential solutions. Having implemented one of the possible solutions, Dynamic Hierarchical Justification, its effectiveness is demonstrated with an empirical analysis.

2M1938+4603 (KIC 9472174) displays a spectacular light curve dominated by a strong reflection effect and rather shallow, grazing eclipses. The orbital period is 0.126 days, the second longest period yet found for an eclipsing sdB+dM, but still close to the minimum 0.1-d period among such systems. The phase-folded light curve was used to detrend the orbital effects from the dataset, and the resulting amplitude spectrum shows a rich collection of pulsation peaks spanning frequencies from ~50 to 4500 uHz. The presence of a complex pulsation spectrum in both the p-mode and the g-mode regions has never been seen before in a compact pulsator. Eclipsing sdB+dM stars are very rare, with only seven systems known and only one with a pulsating primary. Pulsating stars in eclipsing binaries are especially important since they permit masses derived from seismological model fits to be cross checked with orbital mass constraints. We present a first analysis of this star based on the Kepler 9.7-day commissioning light curve and extensive ground-based photometry and spectroscopy that allow us to set useful bounds on the system parameters. We derive a radial-velocity amplitude K_1 = 65.7 +/- 0.6 km/s, inclination angle i = 69.45 +/- 0.20 degrees, and find that the masses of the components are M_1 = 0.48 +/- 0.03 and M_2 = 0.12 +/- 0.01 solar masses., Comment: 5 pages with 3 figures, and 4 appendix pages with 2 figures and 3 tables, submitted to MNRAS

We present new photometric and spectroscopic observations for 2M 1533+3759 (= NSVS 07826147). It has an orbital period of 0.16177042 day, significantly longer than the 2.3--3.0 hour periods of the other known eclipsing sdB+dM systems. Spectroscopic analysis of the hot primary yields Teff = 29230 +/- 125 K, log g = 5.58 +/- 0.03 and log N(He)/N(H) = -2.37 +/- 0.05. The sdB velocity amplitude is K1 = 71.1 +/- 1.0 km/s. The only detectable light contribution from the secondary is due to the surprisingly strong reflection effect. Light curve modeling produced several solutions corresponding to different values of the system mass ratio, q(M2/M1), but only one is consistent with a core helium burning star, q=0.301. The orbital inclination is 86.6 degree. The sdB primary mass is M1 = 0.376 +/- 0.055 Msun and its radius is R1 = 0.166 +/- 0.007 Rsun. 2M1533+3759 joins PG0911+456 (and possibly also HS2333+3927) in having an unusually low mass for an sdB star. SdB stars with masses significantly lower than the canonical value of 0.48 Msun, down to as low as 0.30 Msun, were theoretically predicted by Han et al. (2002, 2003), but observational evidence has only recently begun to confirm the existence of such stars. The existence of core helium burning stars with masses lower than 0.40--0.43 Msun implies that at least some sdB progenitors have initial main sequence masses of 1.8--2.0 Msun or more, i.e. they are at least main sequence A stars. The secondary is a main sequence M5 star., Comment: 47 pages, 7 figures

We present an analysis of three years of precision radial velocity measurements of 160 metal-poor stars observed with HIRES on the Keck 1 telescope. We report on variability and long-term velocity trends for each star in our sample. We identify several long-term, low-amplitude radial-velocity variables worthy of follow-up with direct imaging techniques. We place lower limits on the detectable companion mass as a function of orbital period. Our survey would have detected, with a 99.5% confidence level, over 95% of all companions on low-eccentricity orbits with velocity semi-amplitude K > 100 m/s, or M_p*sin(i) > 3.0 M_JUP*(P/yr)^(1/3), for orbital periods P< 3 yr. None of the stars in our sample exhibits radial-velocity variations compatible with the presence of Jovian planets with periods shorter than the survey duration. The resulting average frequency of gas giants orbiting metal-poor dwarfs with -2.0 < [Fe/H] < -0.6 is f_p<0.67% (at the 1-sigma confidence level). We examine the implications of this null result in the context of the observed correlation between the rate of occurrence of giant planets and the metallicity of their main-sequence solar-type stellar hosts. By combining our dataset with the Fischer & Valenti (2005) uniform sample, we confirm that the likelihood of a star to harbor a planet more massive than Jupiter within 2 AU is a steeply rising function of the host's metallicity. However, the data for stars with -1.0 < [Fe/H] < 0.0 are compatible, in a statistical sense, with a constant occurrence rate f_p~1%. Our results can usefully inform theoretical studies of the process of giant planet formation across two orders of magnitude in metallicity., Comment: 59 pages, 7 tables, 8 figures. Accepted for publication in the Astrophysical Journal

We report new spectroscopic and photometric observations of the parent stars of the recently discovered transiting planets TrES-3 and TrES-4. A detailed abundance analysis based on high-resolution spectra yields [Fe/H] $= -0.19\pm 0.08$, $T_\mathrm{eff} = 5650\pm 75$ K, and $\log g = 4.4\pm 0.1$ for TrES-3, and [Fe/H] $= +0.14\pm 0.09$, $T_\mathrm{eff} = 6200\pm 75$ K, and $\log g = 4.0\pm0.1$ for TrES-4. The accuracy of the effective temperatures is supported by a number of independent consistency checks. The spectroscopic orbital solution for TrES-3 is improved with our new radial-velocity measurements of that system, as are the light-curve parameters for both systems based on newly acquired photometry for TrES-3 and a reanalysis of existing photometry for TrES-4. We have redetermined the stellar parameters taking advantage of the strong constraint provided by the light curves in the form of the normalized separation $a/R_\star$ (related to the stellar density) in conjunction with our new temperatures and metallicities. The masses and radii we derive are $M_\star=0.928_{-0.048}^{+0.028} M_{\sun}$,$R_\star = 0.829_{-0.022}^{+0.015} R_{\sun}$, and $M_\star = 1.404_{-0.134}^{+0.066} M_{\sun}$, $R_\star=1.846_{-0.087}^{+0.096} R_{\sun}$ for TrES-3 and TrES-4, respectively. With these revised stellar parameters we obtain improved values for the planetary masses and radii. We find $M_p = 1.910_{-0.080}^{+0.075} M_\mathrm{Jup}$, $R_p=1.336_{-0.036}^{+0.031} R_\mathrm{Jup}$ for TrES-3, and $M_p=0.925 \pm 0.082 M_\mathrm{Jup}$, $R_p=1.783_{-0.086}^{+0.093} R_\mathrm{Jup}$ for TrES-4. We confirm TrES-4 as the planet with the largest radius among the currently known transiting hot Jupiters., Comment: 42 pages, 10 tables, 8 figures. Accepted for publication in the Astrophysical Journal

We summarize the results of two experiments to address important issues related to the correlation between planet frequencies and properties and the metallicity of the hosts. Our results can usefully inform formation, structural, and evolutionary models of gas giant planets., Comment: 2 pages, no figures. To appear in the proceedings of "IAU conference 249: Exoplanets: Detection, Formation and Dynamics", held in Suzhou, China, 22-26 Oct. 2007

We report on a spectroscopic determination of the atmospheric parameters and chemical abundance of the parent star of the recently discovered transiting planet {TrES-2}. A detailed LTE analysis of a set of \ion{Fe}{1} and \ion{Fe}{2} lines from our Keck spectra yields $T_\mathrm{eff} = 5850\pm 50$ K, $\log g = 4.4\pm 0.1$, and [Fe/H] $= -0.15\pm 0.10$. Several independent checks (e.g., additional spectroscopy, line-depth ratios) confirm the reliability of our spectroscopic $T_\mathrm{eff}$ estimate. The mass and radius of the star, needed to determine the properties of the planet, are traditionally inferred by comparison with stellar evolution models using $T_\mathrm{eff}$ and some measure of the stellar luminosity, such as the spectroscopic surface gravity (when a trigonometric parallax is unavailable, as in this case). We apply here a new method in which we use instead of $\log g$ the normalized separation $a/R_\star$ (related to the stellar density), which can be determined directly from the light curves of transiting planets with much greater precision. With the $a/R_\star$ value from the light curve analysis of Holman et al. \citeyearpar{holman07b} and our $T_\mathrm{eff}$ estimate we obtain $M_\star = 0.980\pm0.062 M_\odot$ and $R_\star = 1.000_{-0.033}^{+0.036} R_\odot$, and an evolutionary age of $5.1^{+2.7}_{-2.3}$ Gyr, in good agreement with other constraints based on the strength of the emission in the \ion{Ca}{2} H & K line cores, the Lithium abundance, and rotation. The new stellar parameters yield improved values for the planetary mass and radius of $M_p = 1.198 \pm 0.053 M_\mathrm{Jup}$ and $R_p = 1.220^{+0.045}_{-0.042} R_\mathrm{Jup}$, confirming that {TrES-2} is the most massive among the currently known nearby ($d\lesssim 300$ pc) transiting hot Jupiters. [Abridged], Comment: 27 pages, 5 figures, 1 table. Accepted for publication in the Astrophysical Journal

We describe a high-precision Doppler search for giant planets orbiting a well-defined sample of metal-poor dwarfs in the field. This experiment constitutes a fundamental test of theoretical predictions which will help discriminate between proposed giant planet formation and migration models. We present here details on the survey as well as an overall assessment of the quality of our measurements, making use of the results for the stars that show no significant velocity variation., Comment: 25 pages, 7 figures, accepted for publication in the Astrophysical Journal

We present a detailed chemical abundance analysis of the parent star of the transiting extrasolar planet TrES-1. Based on high-resolution Keck/HIRES and HET/HRS spectra, we have determined abundances relative to the Sun for 16 elements (Na, Mg, Al, Si, Ca, Sc, Ti, V, Cr, Mn, Co, Ni, Cu, Zn, Y, and Ba). The resulting average abundance of $<[$$X$/H$]>$ $= -0.02\pm0.06$ is in good agreement with initial estimates of solar metallicity based on iron. We compare the elemental abundances of TrES-1 with those of the sample of stars with planets, searching for possible chemical abundance anomalies. TrES-1 appears not to be chemically peculiar in any measurable way. We investigate possible signs of selective accretion of refractory elements in TrES-1 and other stars with planets, and find no statistically significant trends of metallicity [$X$/H] with condensation temperature $T_c$. We use published abundances and kinematic information for the sample of planet-hosting stars (including TrES-1) and several statistical indicators to provide an updated classification in terms of their likelihood to belong to either the thin disk or the thick disk of the Milky Way Galaxy. TrES-1 is found to be a very likely member of the thin disk population. By comparing $\alpha$-element abundances of planet hosts and a large control sample of field stars, we also find that metal-rich ([Fe/H]$\gtrsim 0.0$) stars with planets appear to be systematically underabundant in [$\alpha$/Fe] by $\approx 0.1$ dex with respect to comparison field stars. The reason for this signature is unclear, but systematic differences in the analysis procedures adopted by different groups cannot be ruled out., Comment: 52 pages, 15 figures, 5 tables, accepted for publication in the Astronomical Journal

We present results from our ongoing spectroscopic search for giant planets within 1 AU around a well-defined sample of metal-poor stars with HIRES on the Keck 1 telescope. We have achieved an rms radial velocity precision of $\sim 8$ m/s over a time-span of 1.5 years. The data collected so far build toward evidence of the absence of very short-period ($< 1$ month) giant planets. However, about 7% of the stars in our sample exhibits velocity trends indicative of the existence of companions. We place preliminary upper limits on the detectable companion mass as a function of orbital period, and compare them with the performance of ESA's future space-borne high-precision astrometric observatory Gaia., Comment: 4 pages, 5 figures, Proceedings of the Symposium "The Three-dimensional Universe with Gaia", 4-7 October 2004, Observatoire de Paris-Meudon, France (ESA SP-576)

We report on a spectroscopic determination of the stellar parameters and chemical abundances for the parent star of the transiting planet TrES-1. Based on a detailed analysis of iron lines in our Keck and HET spectra we derive $T_\mathrm{eff} = 5250\pm 75$ K, $\log g = 4.6\pm 0.2$, and [Fe/H] $= 0.00\pm 0.09$. By measuring the \ion{Ca}{2} activity indicator and by putting useful upper limits on the Li abundance we constrain the age of TrES-1 to be $2.5\pm 1.5$ Gyr. By comparing theoretical stellar evolution models with the observational parameters we obtain $M_\star = 0.89\pm 0.05$ $M_\odot$, and $R_\star = 0.83\pm 0.05$ $R_\odot$. Our improved estimates of the stellar parameters are utilized in a new analysis of the transit photometry of TrES-1 to derive a mass $M_p = 0.76\pm 0.05$ $M_\mathrm{J}$, a radius $R_p = 1.04^{+0.08}_{-0.05}$ $R_\mathrm{J}$, and an inclination $i = 89^\circ.5^{+0.5}_{-1.3}$. The improved planetary mass and radius estimates provide the grounds for new crucial tests of theoretical models of evolution and evaporation of irradiated extrasolar giant planets., Comment: 14 pages, 3 figures, Astrophysical Journal Letters, accepted

We present preliminary results from our spectroscopic search for planets within 1 AU of metal-poor field dwarfs using NASA time with HIRES on Keck I. The core accretion model of gas giant planet formation is sensitive to the metallicity of the raw material, while the disk instability model is not. By observing metal-poor stars in the field we eliminate the role of dynamical interactions in dense stellar environments, such as a globular cluster. The results of our survey should allow us to distinguish the relative roles of the two competing giant planet formation scenarios., Comment: 2 pages, to appear in the proceedings of the XIXth IAP Symposium - Extrasolar Planets: Today and Tomorrow, ASP Conference Series

We report a spectroscopic orbit with period P = 3.52433 +/- 0.00027 days for the planetary companion that transits the solar-type star HD209458. For the metallicity, mass, and radius of the star we derive [Fe/H] = 0.00 +/- 0.02, M = 1.1 +/- 0.1 solar masses, and R = 1.3 +/- 0.1 solar radii. This is based on a new analysis of the iron lines in our HIRES template spectrum, and also on the absolute magnitude and color of the star, and uses isochrones from four different sets of stellar evolution models. Using these values for the stellar parameters we reanalyze the transit data and derive an orbital inclination of i = 85.2 +/- 1.4 degrees. For the planet we derive a mass of Mp = 0.69 +/- 0.05 Jupiter masses, a radius of Rp = 1.54 +/- 0.18 Jupiter radii, and a density of 0.23 +/- 0.08 grams per cubic cm., Comment: 11 pages, 1 figure, 2 tables, LaTex, aastex, accepted for publication by ApJ Letters

This paper presents an approach to learning from situated, interactive tutorial instruction within an ongoing agent. Tutorial instruction is a flexible (and thus powerful) paradigm for teaching tasks because it allows an instructor to communicate whatever types of knowledge an agent might need in whatever situations might arise. To support this flexibility, however, the agent must be able to learn multiple kinds of knowledge from a broad range of instructional interactions. Our approach, called situated explanation, achieves such learning through a combination of analytic and inductive techniques. It combines a form of explanation-based learning that is situated for each instruction with a full suite of contextually guided responses to incomplete explanations. The approach is implemented in an agent called Instructo-Soar that learns hierarchies of new tasks and other domain knowledge from interactive natural language instructions. Instructo-Soar meets three key requirements of flexible instructability that distinguish it from previous systems: (1) it can take known or unknown commands at any instruction point; (2) it can handle instructions that apply to either its current situation or to a hypothetical situation specified in language (as in, for instance, conditional instructions); and (3) it can learn, from instructions, each class of knowledge it uses to perform tasks., Comment: See http://www.jair.org/ for any accompanying files